In any traditional network, there is an implicit assumption that flow is conserved on every arc. In generalized networks, each arc has a positive multiplier ?(u, v) called a gain factor, associated with it, representing the fraction of flow that remains when it is sent along that arc. The generalized maximum flow problem is identical to the traditional maximum flow problem, except that it can also model network with “leak” flow. In this paper, we describe the application of the labeling algorithm to the maximumflow generalized network problem. This algorithm has complexity O (EV), where E and V are the number of edges and vertices respectively.